Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2014-02-24

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:01 Mouq Basically the parser pretends its a Perl 6 grammar, taking the .compile_time_value of *everything*
00:01 Mouq I mean...
00:03 Mouq No, I'm overthinking this
00:04 [Coke] colomon: any word from sorear lately
00:04 [Coke] ?
00:05 Mouq Anyway, lue, what exactly would M<> do and how would it fix your problem?
00:06 lue M<backlink:file:line> would do all the work of htmlifying for me, of course :)
00:06 * lue suspects he would be the one to write this module, by the way.
00:06 Mouq Hm.
00:08 Mouq So.. According to spec, that would generate a Pod6::FormattingCode::Name::backlink.new(typename<backlink>, contents<file:line>)
00:08 lue ::Named:: , but yes, that's what it would be.
00:09 Mouq And then....?
00:09 lue https://github.com/perl6/Pod-To-HTML/blob/master/lib/Pod/To/HTML.pm#L385  <--- I wonder if this would handle ::Named::backlink already
00:10 lue content processing would happen within the class, presumably?
00:10 Mouq That's what I don't get
00:10 Mouq Does the class have to custom handle each case of $?DOC ?
00:11 lue I think that function I linked to needs to explicitly support the M code, but that's it for ::HTML.
00:11 Mouq Is the class called by Pod::To::* ?
00:11 lue $?DOC ?
00:11 Mouq http://perlcabal.org/syn/S26.html#How+Pod+is+parsed+and+processed
00:11 Mouq so perl6 --doc=HTML would have a $?DOC eq 'HTML'
00:12 lue Mouq: M<> (and custom-named blocks) feel to me like a situation where you only use them in specific situations.
00:13 lue so yeah, the class itself would be checking $?DOC
00:15 lue r: =begin pod␤M<ok:hello>␤=end pod␤say $=pod;
00:15 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«Pod::Block::Named("name" => "pod").hash␤  Pod::Block::Para␤␤    Pod::FormattingCode("type" => "M").hash␤      ok:hello␤␤␤»
00:15 perigrin joined #perl6
00:15 lue looks like we need some support for M<> in rakudo
00:15 Mouq mmhm
00:16 pippo joined #perl6
00:16 Mouq What if we did this:
00:18 * lue suspects DOC use might be easier to implement than DOC <phaser>, but is only guessing
00:18 Mouq =config M<> { my ($file,$line) = .split(':'); qq[<a href='…] }
00:19 lue would that work in current rakudo?
00:19 Mouq Because there's this whole chain of APIs that we'd need to construct this feature
00:19 Mouq lue: Not right now :P
00:19 Mouq I'm guessing it would parse-fail
00:20 lue I should note that I'm just fine doing it the way the P5 generator does it, no need to make M<> and related work RIGHT NOW or anything :)
00:21 Mouq lue: What it would require is that instead of building the Perl 6 Pod::* objects in NQP, we build ASTs that build it for us
00:21 Mouq Which would simplify everything dramatically, I think
00:22 lue ASTs are likely the way to go when you go beyond simple document markup :)
00:24 Mouq But it requires some serious thought
00:25 lue My two remaining goals for syngen are to place backlinks on a more granular level (i.e. interpret fragments and use them), and dramatically speeding up the HTML generation process
00:25 Mouq lue: How well would my cheat work for you?
00:26 Mouq r: say $=pod.perl;␤␤=pod abcd
00:26 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«Array.new(Pod::Block::Named.new(name => "pod", config => ().hash, content => Array.new(Pod::Block::Para.new(config => ().hash, content => Array.new("abcd")))))␤»
00:26 Mouq ^^ should that work?
00:26 Mouq I mean, conceptually
00:27 Mouq That $=pod gets the Pod after it, because the documents Pod is constructed compile-time
00:27 lue I don't know; it's currently a <div class="smartlink"> surrounding a <p class="smartlink-file">
00:28 lue (the div should contain multiple p elements, though I haven't bothered getting that working yet, if ever)
00:29 lue a Pod6 way of generating the backlink list might make the multiple <p> per <div> easier though :D
00:29 lue =for backlinks\n=backlink L<>\n=backlink ...
00:30 Mouq Hm
00:30 Mouq "It'd be really helpful to have a module providing a generic ay to walk a Pod tree and invoke callbacks on each node"
00:31 lue (then again, detecting when that multiple backlinks per paragraph thing happens should be equally complex with Pod6 constructs or PUA unicode chars...)
00:32 dayangkun joined #perl6
00:32 Mouq lue: With my cheat it's just =begin pod :class<smartlink>␤=pod :class<smartlink-file> L<linkylink>␤=pod :class<smartlink-file>␤=end pod
00:33 Mouq and then the css adds display: block; to .smartlink
00:33 lue yeah, that's essentially what I wrote, except with less syntax sugar :)
00:33 * lue is more concerned with the Pod->HTML chokepoint at the moment :/
00:34 Mouq lue: Well, true, but mine's more generalizable :P and I think that the Pod conversion should be easier to extend
00:34 lue yeah, :class<> would be way more useful, at least in ::HTML land
00:35 Mouq So that you could do `perl6 --doc=HTML,CustomHTML file.pod`
00:36 Mouq And the results of Pod::To::HTML and Pod::To::CustomHTML could be merged. It would just require setting up some kind of API
00:36 Mouq like I said
00:36 Mouq :P
00:36 * lue is trying to see how https://github.com/perl6/doc/blob/master/htmlify.pl does inline HTML conversion
00:39 pdcawley_ joined #perl6
00:42 lue yech, htmlify does too much to the $pod it constructs for me to see clearly what's going on.
00:42 Mouq Huh. Pod::To::HTML can display images written '=Image http://i.imgur.com/YrZrpZj.jpg'
00:43 lue yeah, just noticed that :) .
00:44 lue That means if/when I get around to writing a sub parameters tutorial thing, I can use a pretty picture at one part instead of ASCII art.
00:46 stevan__ joined #perl6
00:48 lee_ joined #perl6
00:49 pdcawley joined #perl6
00:49 lue Come to think of it, when I last ran htmlify it wasn't much faster at making HTML docs that my way... so except for those using my thing with perl6-j, it appears to be more a rakudo/Pod::To::HTML issue.
00:50 dayangkun joined #perl6
00:50 ruoso joined #perl6
00:50 Mouq joined #perl6
00:50 [Sno] joined #perl6
00:50 yihshih joined #perl6
00:50 orevdiabl joined #perl6
00:50 go|dfish joined #perl6
00:52 mavcunha joined #perl6
00:54 Mouq lue: Would this work for you? https://gist.github.com/Mouq/9179799
00:54 colomon [Coke]: I haven't heard anything from sorear++ recently.
00:54 Mouq Aw, lue got netsplit
00:55 Mouq lue: Maybe?
00:56 lue Mouq: no, *you* got netsplit! :)
00:56 Mouq lue: Fair enough :)
00:57 lue looks like it'd work, though I like being able to specify I want a div tag or a p tag .oO(=for Div ?)
00:58 lue <{$node.config<tag>} class="{$node.config<class>}">    :P
00:58 Mouq lue: I don't really feel like that's in the spirit of Pod..
01:00 lue I feel like being able to at least choose div or span, or maybe using div instead of span (this ain't an inline block element in Pod6, after all) would be better that just span
01:00 lue s/that just/than just/
01:01 Mouq lue: Span is as un-opinionated as it gets, and you can always use CSS to make it act like a div
01:01 Mouq I'm trying to keep as much of this out of the actual Pod as possible
01:02 lue Seems like this could require more than just us two discussing :) . For now I'm fine with doing it the way that I have it, though eventually getting rid of an entire stage of the process would be nice.
01:03 * lue is, again, more concerned with the startling lack of speed in converting Pod to HTML, as that's now the longest stage in the process by far.
01:05 Mouq lue: Yeah…
01:05 lue File::Find::find is another timehog in the process.
01:08 Mouq How does one profile Perl6 code?
01:09 lue with --profile, 2>filename, and kcachegrind
01:10 Mouq raydiak++ doc links fixed :D
01:10 raydiak colomon: I just PR'd you a partial fix for a URI::Escape unicode issue
01:11 rurban joined #perl6
01:11 raydiak unbreak all the things!
01:11 colomon raydiak++
01:11 colomon raydiak: should I trust you?
01:11 raydiak colomon: that's a very abstract question
01:11 Mouq "No profiling support"
01:12 raydiak people you can't trust always say "absolutely" to that question :)
01:12 colomon raydiak: feeling lazy enough tonight to do your pull request without testing.  :)
01:12 lue Mouq: it works on perl6-p at least, don't know about -j and -m
01:12 raydiak colomon: oh, yeah I'm pretty confident in this one
01:12 lue (IIRC -m fails, don't know -j)
01:12 colomon raydiak: done
01:13 colomon raydiak: it looked solid, though I have no real clue what's going on at that line.  :)
01:13 raydiak colomon: it says "if this char is lower ascii", but assumes ordinals > 255 don't exist
01:14 colomon right, figured that was what was wrong with the old version
01:14 * lue afk
01:14 raydiak specifically, it checks the highest bit of an 8-bit byte
01:15 raydiak well, 1-byte char...you know what I mean :)
01:15 colomon raydiak: thanks for pinging me, I see so many pull requests for p6 things I usually just ignore them.
01:16 raydiak Mouq, moritz, anyone else who syncs the html docs to feather: make sure to get a fresh URI::Encode and Pod::To::HTML before htmlifying and syncing again
01:16 colomon actually, as I read it, it was just testing the byte #7, right?  So it would have gotten like half of the higher valued Unicode code points.
01:16 Mouq raydiak: will do
01:16 colomon bit #7, not byte, sigh.
01:17 rurban_ joined #perl6
01:17 Mouq raydiak: Also, do you have all the css etc. you have in doc pushed to github? The font seemed larger after you sync'd
01:18 raydiak colomon: yes, it still would have evaluated correctly for many unicode chars, we were just lucky enough to find a failing one :)
01:19 raydiak Mouq: everything looks to be committed and pushed and so forth
01:19 Mouq raydiak: ok
01:25 raydiak Mouq: possibly links I changed going from purple to blue have tricked your eyes?
01:26 klapperl joined #perl6
01:26 raydiak I just went and looked and said "well, it does kinda look bigger in places", but then realized it was just the darker color of unvisited links adding visual weight
01:29 hoverboard joined #perl6
01:30 Mouq raydiak: Maybe
01:32 raydiak Mouq: if you notice anything definite let me know; I didn't (intentionally) do any design work recently
01:33 raydiak oh, and been meaning to say thanks Mouq++ for all the awesome doc work lately
01:34 Mouq np :)
01:35 Mouq Thank you for your work making it awesomer
01:36 raydiak np :) it's exciting to do things that people actually see
01:37 Mouq I have trouble believing that htmlify used to run quite this slow I just didn't notice
01:38 raydiak you aren't running it under moar, are you?
01:38 Mouq I am
01:39 raydiak idk how it compares w/parrot, but it's at least 1 or 2 orders of magnitude slower than doing it w/jvm
01:39 Mouq Huh... I was switching between them last week and thought it was comparable.
01:40 Mouq But I didn't pay as much attention as I should have/timed it
01:41 raydiak the last time I accidentally started a timed run under moar, I lost patience and ^C'd after 40 minutes
01:41 Mouq :|
01:41 raydiak on the same machine jvm does it in 8
01:41 Mouq Do you have a recent MoarVM?
01:42 raydiak yeah, was from nom within the past week
01:43 raydiak oh, and the vm itself was rebuilt at the same time
01:44 Mouq r: say 'infix:<(|)>' ~~ /<!after \\ > '|'/
01:44 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«「|」␤␤»
01:44 Mouq r: say 'infix:<(\|)>' ~~ /<!after \\ > '|'/
01:44 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«Nil␤»
01:45 Mouq r: say 'infix:<(\|)>|routine/infix:<(|)>.html' ~~ /<!after \\ > '|'/
01:45 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«「|」␤␤»
01:45 Mouq r: say ('infix:<(\|)>|routine/infix:<(|)>.html' ~~ /<!after \\ > '|'/).prematch
01:45 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«infix:<(\|)>␤»
01:49 TimToady <!after \\ > is two-pass parsing
01:50 TimToady what if there's a \\ before the \\
01:50 TimToady ?
01:56 Mouq TimToady: Yeah... thank you
02:02 * Mouq points out that perl6-m --target=ast doesn't work and hasn't for a while
02:02 Mouq or maybe ever
02:08 Mouq TimToady: Do you think my $level = 15;␤=begin item :numbered{$level}␤… should work?
02:08 Mouq *:numbered($level)
02:09 Mouq TimToady: Just at first glance with what you think Perl6 should be capable of
02:13 Mouq Well. Hm
02:14 Mouq Maybe BEGIN my $level = 15; maybe makes more sense
02:16 lue Mouq: http://perlcabal.org/syn/S26.html#Delimited+blocks
02:16 lue "All option keys and values must, of course, be constants since Pod is a specification language, not a programming language. Specifically, option values cannot be closures. See Synopsis 2 for details of the various Perl 6 pair notations."
02:16 Mouq Aww
02:16 Mouq Ok
02:17 Mouq That's probably for the best
02:17 lue makes things easier for rakudo at least :)
02:19 klapperl_ joined #perl6
02:23 Mouq r: my @*ABCD;
02:23 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: ( no output )
02:23 Mouq nqp: my @*ABCD;
02:23 camelia nqp-moarvm, nqp-jvm, nqp-parrot: ( no output )
02:26 fridim_ left #perl6
02:29 lue I'd like to find out what's causing the slowdown in Pod->HTML, but I can't get anything out of kcachegrind :(
02:31 Mouq lue: How fast is syngen with perl6-j?
02:31 lue Haven't checked recently, all I know is perl6-m is ≈1200 seconds. I can try to run it on perl6-j though.
02:32 * Mouq complied perl6-p to be able to use its profiler; has no idea how to read it
02:33 lue you have to use something like kcachegrind
02:33 Mouq lue: My OSX is too new for Valgrind TT_TT
02:34 lue kcachegrind doesn't require valgrind, I don't think...
02:39 Mouq lue: I did push :class<> -> <span> to Pod::To::HTML btw
02:39 lue ok
02:40 Mouq dalek: hi
02:44 * lue wishes he were better at interpreting what kcachegrind was telling him
02:50 lue For some reason I'm join() comes up as the offending function. I'm not sure if I should trust that finding though.
02:55 Mouq lue: Join is gonna be called all the time, that seems accurate
02:56 lue or perhaps it's gimme... I seriously don't know how to read all this :/ I'll leave it alone for now, perhaps someone more knowledgeable can help out later.
03:01 skids joined #perl6
03:13 pdcawley joined #perl6
03:16 Mouq nqp: my @a = ['a','b','c']; for @a.kv -> $idx, $letter { say($idx, $letter) }
03:16 camelia nqp-jvm: OUTPUT«Assignment ("=") not supported in NQP, use ":=" instead at line 2, near " ['a','b',"␤  in panic (gen/jvm/stage2/NQPHLL.nqp:378)␤  in infix:sym<=> (gen/jvm/stage2/NQP.nqp)␤  in !protoregex (gen/jvm/stage2/QRegex.nqp:784)␤  in infix (gen/jvm/stage2/NQPHLL.nqp)…»
03:16 camelia ..nqp-moarvm: OUTPUT«Assignment ("=") not supported in NQP, use ":=" instead at line 2, near " ['a','b',"␤   at gen/moar/stage2/NQPHLL.nqp:369  (/home/p6eval/rakudo-inst-1/languages/nqp/lib/NQPHLL.moarvm:panic:120)␤ from <unknown>:1  (/home/p6eval/rakudo-inst-1/languages/nqp/lib/n…»
03:16 camelia ..nqp-parrot: OUTPUT«Assignment ("=") not supported in NQP, use ":=" instead at line 2, near " ['a','b',"␤current instr.: 'panic' pc 15934 (gen/parrot/stage2/NQPHLL.pir:5941) (gen/parrot/stage2/NQPHLL.nqp:425)␤»
03:16 Mouq nqp: my @a := ['a','b','c']; for @a.kv -> $idx, $letter { say($idx, $letter) }
03:16 camelia nqp-parrot: OUTPUT«Method 'kv' not found for invocant of class 'QRPA'␤current instr.: '' pc 81 ((file unknown):52) (/tmp/tmpfile:1)␤»
03:16 camelia ..nqp-moarvm: OUTPUT«Cannot find method 'kv'␤   at /tmp/tmpfile:1  (<ephemeral file>::40)␤ from gen/moar/stage2/NQPHLL.nqp:1090  (/home/p6eval/rakudo-inst-1/languages/nqp/lib/NQPHLL.moarvm::94)␤ from gen/moar/stage2/NQPHLL.nqp:1076  (/home/p6eval/rakudo-inst-1/languages/nqp/lib/…»
03:16 camelia ..nqp-jvm: OUTPUT«Method 'kv' not found for invocant of class 'NQPArray'␤  in  (/tmp/tmpfile:1)␤  in  (gen/jvm/stage2/NQPHLL.nqp:1099)␤  in eval (gen/jvm/stage2/NQPHLL.nqp:1085)␤  in evalfiles (gen/jvm/stage2/NQPHLL.nqp:1291)␤  in command_eval (gen/jvm/stage2/NQPHLL.nqp:1195…»
03:30 TimToady turns out that @a.kv has to calculate keys first, which calls self.end, which forces finite semantics
03:31 TimToady if we invert all the .kv-ish methods to map the values rather than the keys, they can become properly lazy, and maybe run faster too
03:35 Mouq r: my @a = <a b c d e>; @a X, (my$)++
03:35 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: ( no output )
03:35 Mouq r: my @a = <a b c d e>; say @a X, (my$)++
03:35 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«a 0 b 0 c 0 d 0 e 0␤»
03:35 lue Mouq: what are you working on, ooc?
03:35 Mouq Oh, duh, X isn't blocky
03:36 Mouq lue: I finally convinced myself to work on getting tables to work right
03:36 lue Mouq: by the way, perl6-j on syngen takes 3601.609 seconds.
03:37 Mouq r: my @a = <a b c d e>; say gather for @a { take (state$)++; take $_}
03:37 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«0 a 1 b 2 c 3 d 4 e␤»
03:37 Mouq lue: Aww. And Moar++ :)
03:38 lue I think if I were using eval-server or whatever it might run faster...
03:39 Mouq lue: Are you still "$*EXECUTABLE …"?
03:39 lue yes, it runs whatever you ran the main script with.
03:40 Mouq I really don't know how to get around that... I imagine in that case the eval-server might help a lot
03:40 rurban joined #perl6
03:41 lue If Pod::To::HTML came with a way to parse pod files in-script instead of through shell(), then it would definitely be faster for JVM.
03:42 Mouq lue: Pod::To::HTML::pod2html?
03:42 lue don't you need to build up your own object (as I understand from doc/htmlify.pl)?
03:44 Mouq You have to convice perl6 to parse whatever you're looking at as Pod and put it in $=pod (which is a compile-time variable...)
03:44 lue I'd much prefer if there was a HTML::podfile or something
03:44 Mouq s/compile-time variable/variable that has to be known at compile-time/
03:45 lue seems like the kind of thing that's not unreasonable for the module to do instead of me. :)
03:48 Mouq lue: I don't think it would be too difficult to take the Pod stuff from Rakudo and turn it into a Perl 6 grammar+actions, if that's what you mean
03:48 klapperl joined #perl6
03:49 Mouq That could actually be pretty useful
03:49 Mouq But none of the Pod things except <colonpair> reach back out into the rest of the Grammar AFAIK
03:50 lue :/ I'd love to avoid code duplication like that though. It'd preferable to do something like  STD.parsefile($file, :rule<pod_doc>), though I think we're quite a ways away from that.
04:00 Mouq r: DOC use ABCD
04:00 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: ( no output )
04:03 kurahaupo_mobile joined #perl6
04:08 dalek rakudo/nom: e8cea1d | larry++ | src/core/List.pm:
04:08 dalek rakudo/nom: List.end Considered Harmful
04:08 dalek rakudo/nom:
04:08 dalek rakudo/nom: All the .pairs-like list methods calculated .keys first, then looked
04:08 dalek rakudo/nom: up everything using at_pos.  Unfortunately, to determine the key list,
04:08 dalek rakudo/nom: the .keys method called self.end, which enforces finite semantics.  This
04:08 dalek rakudo/nom: seems a bit suboptimal in a language that professes to support infinite
04:08 dalek rakudo/nom: lists.  So now instead of mapping the keys, we map the values, and create
04:08 dalek rakudo/nom: the keys as needed.  Performance seems roughly the same.
04:08 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/e8cea1d5e2
04:09 hoverboard joined #perl6
04:10 TimToady actually, the main difference is the performance on lists that are big but not known to be infinite
04:10 lue that difference is for the better, right? :)
04:10 TimToady for ('a'..'zzzzzz').pairs -> $p {...} no longer hangs forever precalculating the list
04:10 TimToady that's the big win
04:11 kaleem joined #perl6
04:11 Mouq TimToady++ awsummm
04:11 * TimToady should hunt down the rest of the .end calls to see if any of them make sense
04:13 TimToady r: .say for ('a'..'zzzzzz').pairs
04:13 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«(timeout)»
04:14 TimToady in an hour, that should return the initial values
04:17 Mouq r: say ('a'..'zzzzzz')[40]
04:17 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«ao␤»
04:18 Mouq r: say ('a'..'zzzzzz')[40]:delete
04:18 camelia rakudo-parrot 1b5b32: OUTPUT«Can not remove elements from a Range␤  in method gist at gen/parrot/CORE.setting:12260␤  in method gist at gen/parrot/CORE.setting:1042␤  in sub say at gen/parrot/CORE.setting:13157␤  in block  at /tmp/tmpfile:1␤␤»
04:18 camelia ..rakudo-moar 1b5b32: OUTPUT«===SORRY!===␤Can not remove elements from a Range␤»
04:18 camelia ..rakudo-jvm 1b5b32: OUTPUT«Can not remove elements from a Range␤  in method gist at gen/jvm/CORE.setting:12239␤  in method gist at gen/jvm/CORE.setting:1039␤  in sub say at gen/jvm/CORE.setting:13118␤  in block  at /tmp/tmpfile:1␤␤»
04:18 Mouq r: say Array.new('a'..'zzzzzz')[40]
04:18 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«ao␤»
04:18 Mouq r: say Array.new('a'..'zzzzzz')[40]:delete
04:19 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«(timeout)»
04:25 kurahaupo_mobile joined #perl6
04:41 Mouq r: .say for ('a'..'zzzzzz').reduce({$^a,$^b})
04:41 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«(timeout)»
04:49 BenGoldberg rn: for (gather { .take for 1..10; say 'end' }) { .say; last }; say 'after'
04:50 camelia rakudo-jvm 1b5b32: OUTPUT«(timeout)»
04:50 camelia ..rakudo-parrot 1b5b32, rakudo-moar 1b5b32, niecza v24-109-g48a8de3: OUTPUT«1␤after␤»
04:50 BenGoldberg \o/
04:51 * TimToady bows
04:51 BenGoldberg TimToady++
04:52 rurban joined #perl6
04:52 BenGoldberg j: say 'alive'
04:52 camelia rakudo-jvm 1b5b32: OUTPUT«alive␤»
04:52 * BenGoldberg wonders why it timed out on jakudo...
04:54 Mouq rn: for (gather { .take for 1..10; say 'end' }) { .say; last }; say 'after'
04:54 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32, niecza v24-109-g48a8de3: OUTPUT«1␤after␤»
04:57 hoverboard joined #perl6
04:57 BenGoldberg rn: print do { ((state $ = 'a') ^= 'b') ?? 'foo' !! 'bar' }
04:57 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«bar»
04:57 camelia ..niecza v24-109-g48a8de3: OUTPUT«Stack overflow: IP: 0x50322a, fault addr: (nil)␤Stacktrace:␤  at Niecza.Kernel.UnboxAny<T> (Niecza.P6any) <0x0002b>␤  <...>␤  at Niecza.CtxJunctionBool.Get (Niecza.Variable) <0x00117>␤  at Niecza.CtxJunctionBool.Get (Niecza.Variable) <0x0025…»
04:58 * Mouq still doesn't understand what you're trying to accomplish with ^=
04:59 lue it's not updated yet though, is it?
04:59 lue r: .say for ('a'..'zzzzzz').pairs
04:59 Mouq lue: nope
05:00 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«(timeout)»
05:00 * BenGoldberg was just golfing down a bug to the shortest line that exhibits it.
05:00 BenGoldberg n: print do { ((state $ = 'a') |= 'b') ?? 'foo' !! 'bar' }
05:00 camelia niecza v24-109-g48a8de3: OUTPUT«Stack overflow: IP: 0x56fd20, fault addr: (nil)␤Stacktrace:␤  at Niecza.Kernel.UnboxAny<T> (Niecza.P6any) <0x0002b>␤  <...>␤  at Niecza.CtxJunctionBool.Get (Niecza.Variable) <0x00117>␤  at Niecza.CtxJunctionBool.Get (Niecza.Variable) <0x002b…»
05:01 Mouq n: (state $ = False) | True
05:02 camelia niecza v24-109-g48a8de3: ( no output )
05:02 Mouq n: say so (state $ = False) | True
05:02 camelia niecza v24-109-g48a8de3: OUTPUT«True␤»
05:02 Mouq n: (state $ = False) |= True
05:02 camelia niecza v24-109-g48a8de3: ( no output )
05:02 Mouq n: say so ((state $ = False) |= True)
05:02 camelia niecza v24-109-g48a8de3: OUTPUT«Stack overflow: IP: 0x40797a86, fault addr: (nil)␤Stacktrace:␤  at Niecza.Kernel.UnboxAny<T> (Niecza.P6any) <0x0002b>␤  <...>␤  at Niecza.CtxJunctionBool.Get (Niecza.Variable) <0x00117>␤  at Niecza.CtxJunctionBool.Get (Niecza.Variable) <0x00…»
05:09 sftp_ joined #perl6
05:22 hoverboard joined #perl6
05:24 Mouq nqp: my@t:=[[1,2,3],[4,5,6],[7,8,9]]; say([].push(@t)[0][1]);
05:24 camelia nqp-moarvm, nqp-jvm: OUTPUT«2␤»
05:24 camelia ..nqp-parrot: OUTPUT«3␤»
05:26 Mouq nqp: my@t:=[[1,2,3],[4,5,6],[7,8,9]]; my @x; @x.push(@t); say(@x[0][1][1]) # much better
05:26 camelia nqp-moarvm, nqp-jvm, nqp-parrot: OUTPUT«5␤»
05:42 TimToady r: .say for ('a'..'zzzzzz').pairs
05:42 camelia rakudo-parrot 1b5b32, rakudo-jvm 1b5b32, rakudo-moar 1b5b32: OUTPUT«(timeout)»
05:50 kurahaupo_mobile joined #perl6
05:55 kurahaupo__ joined #perl6
06:05 kurahaupo_mobile joined #perl6
06:06 BenGoldberg r: ('a'..'zzzzzz').pairs.infinite.say
06:06 camelia rakudo-parrot e8cea1, rakudo-jvm e8cea1, rakudo-moar e8cea1: OUTPUT«Nil␤»
06:06 Mouq Updated :)
06:06 BenGoldberg r: for( ('a'..'zzzzzz').pairs ) { .say; last }
06:06 camelia rakudo-parrot e8cea1, rakudo-jvm e8cea1, rakudo-moar e8cea1: OUTPUT«[31m===[0mSORRY![31m===[0m�Word 'for' interpreted as 'for()' function call; please use whitespace instead of parens�at /tmp/tmpfile:1�------> [32mfor[33m�[31m( ('a'..'zzzzzz').pairs ) { .say; last }[…»
06:06 BenGoldberg r: for ( ('a'..'zzzzzz').pairs ) { .say; last }
06:06 camelia rakudo-parrot e8cea1, rakudo-jvm e8cea1, rakudo-moar e8cea1: OUTPUT«0 => "a"␤»
06:07 BenGoldberg r: for ( ('a'..'zzzzzz').pairs ) { .say; last if $_ ge 'aa' }
06:07 camelia rakudo-moar e8cea1: OUTPUT«(timeout)0 => "a"␤1 => "b"␤2 => "c"␤3 => "d"␤4 => "e"␤5 => "f"␤6 => "g"␤7 => "h"␤8 => "i"␤9 => "j"␤10 => "k"␤11 => "l"␤12 => "m"␤13 => "n"␤14 => "o"␤15 => "p"␤16 => "q"␤17 => "r"␤18 => "s"␤19 => "t"␤20 => "u"␤21 =>…»
06:07 camelia ..rakudo-jvm e8cea1: OUTPUT«(timeout)0 => "a"␤1 => "b"␤2 => "c"␤3 => "d"␤4 => "e"␤5 => "f"␤6 => "g"␤7 => "h"␤8 => "i"␤9 => "j"␤10 => "k"␤11 => "l"␤12 => "m"␤13 => "n"␤14 => "o"␤15 => "p"␤16 => "q"␤17 => "r"␤18 => "s"␤19 => "t"␤20 => "u"␤21 => …»
06:07 camelia ..rakudo-parrot e8cea1: OUTPUT«(timeout)0 => "a"␤1 => "b"␤2 => "c"␤3 => "d"␤4 => "e"␤5 => "f"␤6 => "g"␤7 => "h"␤8 => "i"␤9 => "j"␤10 => "k"␤11 => "l"␤12 => "m"␤13 => "n"␤14 => "o"␤15 => "p"␤16 => "q"␤17 => "r"␤18 => "s"␤19 => "t"␤20 => "u"␤21 …»
06:08 TimToady r: .say for ('a'..'zzzzzz').pairs
06:08 camelia rakudo-parrot e8cea1: OUTPUT«(timeout)0 => "a"␤1 => "b"␤2 => "c"␤3 => "d"␤4 => "e"␤5 => "f"␤6 => "g"␤7 => "h"␤8 => "i"␤9 => "j"␤10 => "k"␤11 => "l"␤12 => "m"␤13 => "n"␤14 => "o"␤15 => "p"␤16 => "q"␤17 => "r"␤18 => "s"␤19 => "t"␤20 => "u"␤21 …»
06:08 camelia ..rakudo-jvm e8cea1: OUTPUT«(timeout)0 => "a"␤1 => "b"␤2 => "c"␤3 => "d"␤4 => "e"␤5 => "f"␤6 => "g"␤7 => "h"␤8 => "i"␤9 => "j"␤10 => "k"␤11 => "l"␤12 => "m"␤13 => "n"␤14 => "o"␤15 => "p"␤16 => "q"␤17 => "r"␤18 => "s"␤19 => "t"␤20 => "u"␤21 => …»
06:08 camelia ..rakudo-moar e8cea1: OUTPUT«(timeout)0 => "a"␤1 => "b"␤2 => "c"␤3 => "d"␤4 => "e"␤5 => "f"␤6 => "g"␤7 => "h"␤8 => "i"␤9 => "j"␤10 => "k"␤11 => "l"␤12 => "m"␤13 => "n"␤14 => "o"␤15 => "p"␤16 => "q"␤17 => "r"␤18 => "s"␤19 => "t"␤20 => "u"␤21 =>…»
06:11 BenGoldberg r: for ( ('a'..'zzzzzz').pairs ) { .say; last if .value eq 'aa' }
06:11 TimToady r: for ('a'..'zzzzzz').pairs { .say; last if .value ge 'aa' }
06:11 BenGoldberg :)
06:11 camelia rakudo-parrot e8cea1, rakudo-jvm e8cea1, rakudo-moar e8cea1: OUTPUT«0 => "a"␤1 => "b"␤2 => "c"␤3 => "d"␤4 => "e"␤5 => "f"␤6 => "g"␤7 => "h"␤8 => "i"␤9 => "j"␤10 => "k"␤11 => "l"␤12 => "m"␤13 => "n"␤14 => "o"␤15 => "p"␤16 => "q"␤17 => "r"␤18 => "s"…»
06:11 camelia rakudo-parrot e8cea1, rakudo-jvm e8cea1, rakudo-moar e8cea1: OUTPUT«0 => "a"␤1 => "b"␤»
06:11 TimToady 'b' ge 'aa' :)
06:12 TimToady and you don't need the supersitious parens :P
06:13 TimToady a method doesn't require (), unlike, say, lines()
06:14 SamuraiJack_ joined #perl6
06:21 darutoko joined #perl6
06:25 berekuk joined #perl6
06:55 [Sno] joined #perl6
06:56 Mouq lue: "Blocks whose names contain at least one uppercase and one lowercase letter are assumed to be destined for specialized renderers or parser plug-ins." I hadn't seen this, reconsidering "=for Div" etc.
07:02 kurahaupo_mobile joined #perl6
07:03 dmol joined #perl6
07:10 rurban joined #perl6
07:15 aborazmeh joined #perl6
07:21 FROGGS joined #perl6
07:30 FROGGS masak: do you think this should be specced to work? my regex foo { aaa }; say "aaa" ~~ foo # perhaps same for grammars
07:36 TimToady r: my regex foo { aaa }; say "aaa" ~~ &foo
07:36 camelia rakudo-jvm e8cea1: OUTPUT«(timeout)»
07:36 camelia ..rakudo-parrot e8cea1, rakudo-moar e8cea1: OUTPUT«「aaa」␤␤»
07:36 TimToady you need the noun form, not the verb form
07:38 TimToady otherwise it's going to try to call foo() first, then do the ~~
07:41 [Coke] std: my $s = 'abcd'; while $s ~~ m:g/(.)/ { print $0 };
07:41 camelia std 09dda5b: OUTPUT«ok 00:01 129m␤»
07:41 FROGGS TimToady++
07:41 [Coke] ^^ RT #77144 is a request for that to warn.
07:41 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=77144
07:41 FROGGS masak: that is what you want perhaps :o) ----- ^
07:42 FROGGS (not the RT link though)
07:42 berekuk joined #perl6
07:51 TimToady [Coke]: detecting list/scalar context confusion is going to be hard in a language that handles context lazily
07:52 [Coke] Then I'll reject the ticket, hokay?
07:52 [Coke] Here's another ticket that requires a ruling, and perhaps a rethink in the 3.5 years since it was opened: RT #77340
07:53 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=77340
07:53 Quom joined #perl6
07:54 robinsmidsrod joined #perl6
07:57 sftp joined #perl6
07:58 yihshih__ joined #perl6
07:58 TimToady given current understanding of how things itemize, I think Pm has it right, and we should reject the ticket
07:58 TimToady except that @%h<abc> can't work
07:58 ruoso_ joined #perl6
07:58 TimToady must be either @(%h<abc>) or %h<abc>.list or %h<abc>[]
07:59 [Coke] TimToady++
07:59 pdcawley_ joined #perl6
08:01 [Coke] r: say say $*PERL
08:01 camelia rakudo-moar e8cea1: OUTPUT«("name" => "rakudo", "compiler" => {"name" => "rakudo", "ver" => "2014.02-18-ge8cea1d", "release-number" => "", "build-date" => "2014-02-24T06:25:33Z", "codename" => ""}).hash␤True␤»
08:01 camelia ..rakudo-parrot e8cea1: OUTPUT«("name" => "rakudo", "compiler" => {"name" => "rakudo", "ver" => "2014.02-18-ge8cea1d", "release-number" => "", "build-date" => "2014-02-24T06:32:25Z", "codename" => ""}).hash␤True␤»
08:01 camelia ..rakudo-jvm e8cea1: OUTPUT«("name" => "rakudo", "compiler" => {"codename" => "", "name" => "rakudo", "release-number" => "", "ver" => "2014.02-18-ge8cea1d", "build-date" => "2014-02-24T05:12:38Z"}).hash␤True␤»
08:01 [Coke] r: say $*PERL
08:01 camelia rakudo-parrot e8cea1: OUTPUT«("name" => "rakudo", "compiler" => {"name" => "rakudo", "ver" => "2014.02-18-ge8cea1d", "release-number" => "", "build-date" => "2014-02-24T06:32:25Z", "codename" => ""}).hash␤»
08:01 camelia ..rakudo-moar e8cea1: OUTPUT«("name" => "rakudo", "compiler" => {"name" => "rakudo", "ver" => "2014.02-18-ge8cea1d", "release-number" => "", "build-date" => "2014-02-24T06:25:33Z", "codename" => ""}).hash␤»
08:01 camelia ..rakudo-jvm e8cea1: OUTPUT«("name" => "rakudo", "compiler" => {"codename" => "", "name" => "rakudo", "release-number" => "", "ver" => "2014.02-18-ge8cea1d", "build-date" => "2014-02-24T05:12:38Z"}).hash␤»
08:02 [Coke] urk. I expected that to fail, so resent.
08:02 [Coke] r: say $*PERL<version>
08:02 camelia rakudo-parrot e8cea1, rakudo-jvm e8cea1, rakudo-moar e8cea1: OUTPUT«(Any)␤»
08:04 revdiablo joined #perl6
08:08 [Coke] rakudo: my Int (Str $x); say $x.WHAT # unspecced behavior?
08:08 camelia rakudo-parrot e8cea1, rakudo-jvm e8cea1, rakudo-moar e8cea1: OUTPUT«(Int)␤»
08:09 [Coke] rakudo: sub foo() { return 1,2,:c<3> }; say foo().perl
08:09 camelia rakudo-parrot e8cea1, rakudo-jvm e8cea1, rakudo-moar e8cea1: OUTPUT«(1, 2, "c" => Mu)␤»
08:09 [Coke] rakudo: say [1,2] X~ <a b>
08:09 camelia rakudo-parrot e8cea1, rakudo-jvm e8cea1, rakudo-moar e8cea1: OUTPUT«1a 1b 2a 2b␤»
08:11 [Coke] rakudo: my $*a = 42; foo; sub foo { say $*a; my $*a }
08:11 camelia rakudo-parrot e8cea1, rakudo-jvm e8cea1, rakudo-moar e8cea1: OUTPUT«(Any)␤»
08:11 go|dfish joined #perl6
08:12 FROGGS joined #perl6
08:15 [Coke] nom: class A { eval 'has $.x' }; say A.new(x => 3).x
08:15 camelia rakudo-parrot e8cea1, rakudo-jvm e8cea1: OUTPUT«===SORRY!===␤No such method 'name' for invocant of type 'NQPMu'␤Saw 1 call to deprecated code during execution.␤================================================================================␤Sub eval (from GLOBAL) called at:…»
08:15 camelia ..rakudo-moar e8cea1: OUTPUT«===SORRY!===␤Cannot find method 'name'␤»
08:15 kaleem joined #perl6
08:15 [Coke] nom: class A { EVAL 'has $.x' }; say A.new(x => 3).x
08:15 camelia rakudo-parrot e8cea1, rakudo-jvm e8cea1: OUTPUT«===SORRY!===␤No such method 'name' for invocant of type 'NQPMu'␤»
08:15 camelia ..rakudo-moar e8cea1: OUTPUT«===SORRY!===␤Cannot find method 'name'␤»
08:16 [Coke] nom: class A { BEGIN eval 'has $.x' }; say A.new(x => 3).x
08:16 [Coke] nom: class A { BEGIN EVAL 'has $.x' }; say A.new(x => 3).x
08:16 moritz .oO( we still have a 'nom' target? :-)
08:16 camelia rakudo-jvm e8cea1: OUTPUT«(timeout)»
08:16 camelia ..rakudo-parrot e8cea1: OUTPUT«===SORRY!===␤No such method 'name' for invocant of type 'NQPMu'␤Saw 1 call to deprecated code during execution.␤================================================================================␤Sub eval (from GLOBAL) called at:␤  /tmp/tmpfile, l…»
08:16 camelia ..rakudo-moar e8cea1: OUTPUT«===SORRY!===␤Cannot find method 'name'␤»
08:16 camelia rakudo-jvm e8cea1: OUTPUT«(timeout)»
08:16 camelia ..rakudo-parrot e8cea1: OUTPUT«===SORRY!===␤No such method 'name' for invocant of type 'NQPMu'␤»
08:16 camelia ..rakudo-moar e8cea1: OUTPUT«===SORRY!===␤Cannot find method 'name'␤»
08:16 [Coke] nom: has $.x
08:17 camelia rakudo-parrot e8cea1, rakudo-jvm e8cea1, rakudo-moar e8cea1: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/tmpfileâ�¤You cannot declare attribute '$.x' here; maybe you'd like a class or a role?â�¤at /tmp/tmpfile:1â�¤------> [32mhas $.x[33mâ��[31m<EOL>[0mâ�¤   …»
08:18 [Coke] whoops, meant for all that to go to a private window.
08:18 FROGGS :o)
08:19 [Coke] ok, only 83 more tickets that haven't been touched in 2+ years.
08:20 [Coke] up to 34 testneeded tickets - those are often LHF for people looking to contribute. Feel free to grab one and submit a test. (usually the code is in the ticket.)
08:25 FROGGS I have some left in my inbox that I want to do
08:27 dalek rakudo/nom: 299d708 | larry++ | src/core/List.pm:
08:27 dalek rakudo/nom: combinations didn't itemize when k == 0
08:27 dalek rakudo/nom:
08:27 dalek rakudo/nom: The null set was omitted from the powerset.
08:27 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/299d708224
08:28 moritz m: say (^5).list.keys; say (^5).list.keys;
08:28 camelia rakudo-moar e8cea1: OUTPUT«0 1 2 3 4␤0 1 2 3 4␤»
08:31 dalek roast: 2582eed | moritz++ | S32-exceptions/misc.t:
08:31 dalek roast: Tests for RT #67554
08:31 dalek roast: review: https://github.com/perl6/roast/commit/2582eed4dc
08:31 synopsebot Link: https://rt.perl.org/rt3//Public/Bug/Display.html?id=67554
08:35 [Coke] moritz++
08:41 rindolf joined #perl6
08:44 rurban joined #perl6
08:53 salv0 joined #perl6
09:06 denis_boyun joined #perl6
09:09 zakharyas joined #perl6
09:23 Alula joined #perl6
09:31 rindolf joined #perl6
09:33 grondilu joined #perl6
09:34 fhelmberger joined #perl6
09:35 grondilu on http://rosettacode.org/wiki/Almost_prime#Perl_6, TimToady wrote about a factors routine: « (to be included manually until we decide where in the distribution to put it) ».  Couldn't it just be a Int method?
09:40 mavcunha joined #perl6
09:50 mavcunha_ joined #perl6
09:53 dakkar joined #perl6
10:28 pecastro joined #perl6
10:32 Alula joined #perl6
10:35 nwc10 is this known to happen:
10:35 nwc10 rn: say 1.is-prime
10:35 camelia rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70, niecza v24-109-g48a8de3: OUTPUT«False␤»
10:35 nwc10 oh, wait. so something screwed locally. hmmm
10:35 nwc10 live demo fail! :-)
10:37 jnthn Never happens to me :P
10:38 jnthn Comes out False on my local too, fwiw.
10:38 nwc10 I might have inconsistent versions of stuff
10:38 jnthn My local is on moar-conc, which is bleading in different ways than nom :)
10:43 zakharyas joined #perl6
10:44 moritz nwc10: maybe your 1 was twice as large as the usual 1 :-)
10:47 kaleem joined #perl6
10:48 jnthn r: multi infix:<==>(2, 1) { True }; say 2 == 1
10:48 camelia rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70: OUTPUT«True␤»
10:48 moritz jnthn: you are EVAL, erm, evil :-)
10:54 jnthn True evil is sneaking that into a module and adding "is export" :P
10:54 moritz aren't multis supposed to be exported by default anyway? :-)
10:54 jnthn S11 may well argue for that :)
10:55 jnthn It's a little more magic than I like tbh.
10:55 moritz same here
10:55 jnthn As in, it's not unusual to use multis inside a module as an implementation detail.
10:55 nwc10 i
10:56 jnthn nwc10: Most complex message so far today...
10:56 nwc10 yes, and it was meant for mutt
10:57 moritz [Coke]++ just quoted S11 for a ticket
10:57 moritz Any C<proto> declaration that is not declared C<my> is exported by default.
10:57 moritz Any C<multi> that depends on an exported C<proto> is also automatically
10:57 moritz exported.  Any autogenerated C<proto> is assumed to be exported by default.
10:58 moritz are protos in the setting exported automatically?
10:58 moritz if not, adding a multi a setting proto from a module wouldn't export it
10:59 jnthn Well, that's debatable. It sitll has to generate a proto in the nested scope.
10:59 jnthn *still
10:59 jnthn That is, it copies the outer proto with its candidate list and then appends to it.
11:05 breinbaas joined #perl6
11:05 _sri joined #perl6
11:09 grondilu_ joined #perl6
11:10 pecastro_ joined #perl6
11:17 timotimo lue: are you running the pod-to-html stuff on moarvm?
11:17 silug__ joined #perl6
11:18 pdcawley joined #perl6
11:20 bjz joined #perl6
11:20 robinsmidsrod joined #perl6
11:25 * timotimo doesn't have the patience for a full backlog right now :|
11:26 donaldh joined #perl6
11:28 donaldh joined #perl6
11:37 lizmat joined #perl6
11:41 timotimo [Coke]++ # amazing amounts of bug triage
11:48 masak good afternoon, #perl6
11:48 * masak is teaching Python today
11:48 * masak .oO( hssssSSSS! )
11:48 nwc10 heresy! :-)
11:49 masak strangely enough, it feels very little like heresy.
11:50 * jnthn will be teaching lots of C# in March. Doesn't feel like heresy either. :)
11:51 jnthn Though I do keep missing things from Perl 6. :)
11:51 masak think of us as scouts doing forays into other communities :)
11:51 jnthn The .Net folks are oddest generally, in so far as they tend to do .Net stuff and nothing else...
11:52 masak I hear COBOL folks are much the same :P
11:52 jnthn They've too much to write to have time to do anything else :P
11:53 masak bet it pays well, too.
11:53 jnthn Yeah, well. The 2038 bug will pay our generation's retirement well too. :)
11:53 masak "don't you dare die! then who will maintain our legacy bank software?"
11:53 masak jnthn: all we need to do is to create a massive 2038 scare :)
11:53 masak challenge accepted!
11:54 jnthn masak: Easy enough. Bad news sells. :)
11:54 * masak .oO( "Ten things you wouldn't believe about our eventual destruction in the year 2038" )
11:54 jnthn 32 things, surely?
11:55 masak :P
11:55 masak -- signed, int
11:55 masak or is that 'signed, long' ?
11:55 jnthn Platform specific, naturally... :)
11:59 masak :)
12:04 timotimo moritz: would like support for "/today/" and "/yesterday/" as well as those"/summary/" URLs in the irclogs
12:05 donaldh today works, no?
12:06 jnthn yeah, today already does.
12:06 timotimo not in combination with /summary/
12:07 donaldh ah :)
12:10 moritz timotimo: https://github.com/moritz/ilbot/issues/33 same problem with /text/
12:11 moritz timotimo: maybe add it to the ticket? (or even pull-request? :-)
12:12 timotimo thank you
12:12 moritz (all URLs are handled in ilbot.psgi)
12:13 timotimo i'd like the CSRF error for the rakudo.org/tickets links to go away
12:15 moritz CSRF error?
12:15 donaldh oh, panda bootstrap fails on JVM on OS X. Is it known to work / not work on JVM on other platforms?
12:15 timotimo RT has detected a possible cross-site request forgery for this request, because the Referrer header supplied by your browser (rakudo.org:80) is not allowed by RT's configured hostname (rt.perl.org:443) or whitelisted hosts (www.bitcard.org:443, www.bestpractical.com:443). A malicious attacker may be trying to modify or access a search on your behalf. If you did not initiate this request, then you should alert
12:15 timotimo your security team.
12:15 moritz eeks
12:16 moritz oh, and some of the search URLs produce other RT errors
12:23 timotimo http://p6weekly.wordpress.com/?p=68&amp;shareadraft=530b399b58b78 - first draft of today's p6weekly post
12:23 nwc10 donaldh: not sure. hoelzro I think was working on it
12:25 moritz timotimo: not super exciting, but there are some new docs on doc.perl6.org
12:26 timotimo that's cool. which ones?
12:26 timotimo ah, i'll just look at the commit log
12:26 moritz IO::Handle, IO::FileTestable, Mix, MixHash, Mixy, QuantHash, Set, SetHash, Setty, regexes
12:26 moritz Mouq++ did most of those (I only did regexes)
12:26 timotimo thank you!
12:26 moritz http://doc.perl6.org/language/regexes # ouch, something very wrong with table parsing or rendering
12:28 jnthn timotimo: "The most noticeable thing still missing is ThreadPool and its Scheduler."
12:28 jnthn timotimo: Actually ThreadPoolScheduler is working now, for non-time-based scheduling.
12:28 timotimo oh!
12:28 jnthn timotimo: So we're onto Promise itself.
12:28 timotimo excellent!
12:28 timotimo thanks for the correction
12:29 jnthn timotimo: Also, didn't you contribute something to this effort as well as me and JimmyZ? ;)
12:29 timotimo only tiny bits
12:29 * timotimo doesn't like tooting his own horn too much :P
12:29 jnthn Still helpful bits :)
12:31 jnthn timotimo: In what's cooking, I expect I'll have promises working on Moar, and likely channels too, during the next week.
12:31 timotimo damnit!
12:31 jnthn timotimo: Probably supplies too. I think not the timer bits, though. :)
12:32 jnthn I know. Bloody progress. :P
12:32 timotimo i just lost most of the text >_<
12:32 timotimo changed from "Text" to "Visual" and hit ctrl-z
12:32 timotimo very much not recommended at all
12:32 nwc10 it's the old "this week" problem :-)
12:33 jnthn timotimo: I still have it in a tab here?
12:33 jnthn timotimo: Do you want the text or the HTML?
12:33 timotimo i still have that tab, too
12:33 jnthn ah, ok :)
12:33 timotimo i was able to copy the text + formatting into the wysiwyg editor
12:34 * jnthn figures its lunch time...
12:34 timotimo bon appetit :)
12:40 timotimo http://p6weekly.wordpress.com/?p=68&amp;shareadraft=530b3d9e6bc6d - updated with hopefully all corrections
12:42 timotimo http://p6weekly.wordpress.com/?p=68&amp;shareadraft=530b3e4a495bf - more changes
12:43 jnthn "Mouq and lue have been working on Pod6 support" - may be worth saying "working on improving Pod6 support", so it's clear we aren't just getting started on it :)
12:44 moritz timotimo: regexes.pod is meant as reference more than a tutorial
12:44 timotimo ah, yes!
12:44 timotimo i had that change before i lost my stuff
12:44 moritz (I started of writing it as a tutorial, but it becomes more of a ref now)
12:45 grondilu_ std: my $x = 1; say $x*$x<=1
12:45 camelia std 09dda5b: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Unable to parse quote-words subscript; couldn't find right angle quote at /tmp/Ek__fzleho line 1:â�¤------> [32mmy $x = 1; say $x*$x<[33mâ��[31m=1[0mâ�¤    expecting escapeâ�¤Parse failedâ�¤FAILED 00:01 125mâ�¤Â»
12:45 grondilu_ :/
12:46 grondilu joined #perl6
12:46 grondilu std: my $x = 1; say $x*$x <= 1
12:46 camelia std 09dda5b: OUTPUT«ok 00:01 124m␤»
12:46 grondilu std: my $x = 1; say $x*$x <=1
12:46 camelia std 09dda5b: OUTPUT«ok 00:01 124m␤»
12:46 grondilu std: my $x = 1; say $x*$x<= 1
12:46 camelia std 09dda5b: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Unable to parse quote-words subscript; couldn't find right angle quote at /tmp/91VugElieO line 1:â�¤------> [32mmy $x = 1; say $x*$x<[33mâ��[31m= 1[0mâ�¤    expecting escapeâ�¤Parse failedâ�¤FAILED 00:01 125mâ�¤Â»
12:47 timotimo http://p6weekly.wordpress.com/?p=68&amp;shareadraft=530b3f3db24d3 - most up to date by far!
12:49 * timotimo recommends to click the "save draft" button regularly in wordpress' editor
12:49 masak timotimo: changing to "Visual" is very much not recommended at all. :)
12:50 timotimo aye.
12:50 masak timotimo: I tend to hit "Preview".
12:50 masak WYSIWYG is a big lie, *especially* when there's an underlying textual representation.
12:50 grondilu std: $_*$_<= 1
12:50 camelia std 09dda5b: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Unable to parse quote-words subscript; couldn't find right angle quote at /tmp/7Quo3EsW6h line 1:â�¤------> [32m$_*$_<[33mâ��[31m= 1[0mâ�¤    expecting escapeâ�¤Parse failedâ�¤FAILED 00:01 124mâ�¤Â»
12:50 masak grondilu: whitespace.
12:50 moritz grondilu: "use whitespace around infix operators"
12:51 masak grondilu: if you skip the whitespace before an infix, it *becomes* a postfix or postcircumfix.
12:51 grondilu I thought this was only a best practice, not mandatory
12:51 masak grondilu: it's only best practice
12:51 masak but for a good reason :)
12:52 moritz grondilu: it's a best practise you should follow, because if not, you always need to be aware of potential ambiguties
12:52 timotimo http://p6weekly.wordpress.com/?p=68&amp;shareadraft=530b4074773d7 <- mentioned the lack of rakudo star release manager
12:52 moritz https://github.com/moritz/perl6-wtop/blob/master/practices.pod
12:53 grondilu shouldn't LTM fall back to the only interpretation that actually parses or something?  I mean, there is no > so the < should not be interpreted as the beginning of a postfix, should it?
12:53 timotimo why have i never heard of perl6-wtop before?
12:53 masak grondilu: no, LTM doesn't work like that.
12:53 jnthn grondilu: LTM isn't a magic unicorn :P
12:54 * timotimo notices he doesn't have any bread in the home
12:54 masak grondilu: think of it as being greedy on individual tokens.
12:54 moritz grondilu: that would be backtracking
12:54 jnthn grondilu: It works at *token* level.
12:54 moritz and backtracking usually leads to very unintuitive behavior
12:54 grondilu ok
12:55 tgt joined #perl6
12:55 timotimo moritz: can you use definition lists for the section "backslashed, predefined character classes" perhaps?
12:56 moritz timotimo: I don't know how to write those in pod6
12:56 timotimo OK
12:56 * timotimo is AFK for a bit and would be happy about more feedback to the post
13:01 * timotimo added numeric formatting test suite to the LHF
13:05 daniel-s_ joined #perl6
13:06 denis_boyun joined #perl6
13:12 zakharyas joined #perl6
13:20 btyler joined #perl6
13:28 colomon joined #perl6
13:29 nwc10 timotimo: I think your wording for "You can safely disregard the huge Cross Site Request Forgery warning that may pop up and just “click here to resume your request”" is unfortunate
13:31 nwc10 I think something more exact like "we are aware that you may get a warning about a Cross Site Request Forgery from rakudo.org to rt.perl.org. We're working on how to avoid this happening, but for now please use “click here to resume your request” for this specific warning"
13:32 nwc10 because it reads like you're telling people to ignore that warning always. Which you aren't, but "people are stupid"
13:32 nwc10 or at least, play stupid on TV.
13:32 kst`` joined #perl6
13:33 DrEeevil joined #perl6
13:33 Shozan joined #perl6
13:34 Kelder joined #perl6
13:34 araujo joined #perl6
13:34 araujo joined #perl6
13:35 yath_ joined #perl6
13:35 kshannon_ joined #perl6
13:35 JimmyZ_ joined #perl6
13:35 dagurval_ joined #perl6
13:35 cxreg2 joined #perl6
13:35 segomos_ joined #perl6
13:35 cosimo_ joined #perl6
13:35 BooK_ joined #perl6
13:35 Psyche^_ joined #perl6
13:35 pmichaud_ joined #perl6
13:38 exodist_ joined #perl6
13:39 eternaleye_ joined #perl6
13:40 btyler_ joined #perl6
13:40 genehack_ joined #perl6
13:40 timotimo http://p6weekly.wordpress.com/?p=68&amp;shareadraft=530b4bbec4c9e
13:40 timotimo thanks
13:40 mathw joined #perl6
13:46 woolfy1 joined #perl6
13:46 yogan_ joined #perl6
13:46 geekosaur joined #perl6
13:46 Bucciarati joined #perl6
13:46 atrodo joined #perl6
13:47 avuserow3 joined #perl6
13:47 yves_ joined #perl6
13:48 denis_boyun_ joined #perl6
13:48 _sri joined #perl6
13:48 lue joined #perl6
13:48 Alina-malina joined #perl6
13:49 hoelzro morning, #perl6
13:49 atta joined #perl6
13:49 Khisanth joined #perl6
13:49 ajr joined #perl6
13:51 pdcawley joined #perl6
13:53 masak hoelzro! \o/
13:54 xenoterracide joined #perl6
13:56 nwc10 hoelzro: offhand, do you know the answer to:  oh, panda bootstrap fails on JVM on OS X. Is it known to work / not work on JVM on other platforms?
13:56 nwc10 which was http://irclog.perlgeek.de/perl6/2014-02-24#i_8335189
13:56 hoelzro nwc10: I haven't worked on panda yet
13:56 hoelzro I did work to make ufo work on the different backends, though
13:56 donaldh Ah, panda bootstrap is recursively copying .work when running local-fetch for panda itself.
13:57 nwc10 hoelzro: sorry, I didn't realise that.
13:57 hoelzro np =)
13:57 nwc10 I'm very pleased that people are working on this stuff
13:57 nwc10 (as I've got too much other stuff on the go)
13:57 donaldh Failed to create directory '.work/1393250114_7//.work/1393250114_7/.work/1393250114_7/.work/ ....... : File name too long
13:58 donaldh I'll work on a fix.
13:59 hoelzro donaldh: I think someone reported that dir() returns relative (rather than absolute) paths on various backends
13:59 hoelzro iirc, it's supposed to return absolute
14:00 hoelzro that sounds like it could result in that recursion issue
14:01 moritz iirc it's supposed to return Path objects
14:01 hoelzro timotimo++ # post
14:01 hoelzro moritz: it does
14:01 yoleaux joined #perl6
14:02 hoelzro someone mentioned something about abs vs rel, though
14:08 kbaker_ joined #perl6
14:09 virtualsue joined #perl6
14:09 colomon joined #perl6
14:12 FROGGS donaldh: have a look at the patch in my comment: https://github.com/tadzik/panda/pull/68
14:13 smls joined #perl6
14:13 FROGGS donaldh: and also have a look at this for the .work problem:
14:14 FROGGS https://gist.github.com/Mouq/cd9363ce6f20c699af77
14:14 jnap joined #perl6
14:14 donaldh FROGGS: Thanks. Yeah, I suspected lazy.
14:14 FROGGS donaldh: if you have time to fix it please go ahead
14:17 timotimo https://p6weekly.wordpress.com/2014/02/24/changes-since-release-2014-02/ - published
14:22 JimmyZ_ timotimo++
14:24 * timotimo pokes perlbug-admin again about the CSRF issue
14:24 skids joined #perl6
14:28 smls Hi
14:28 yoleaux 16 Feb 2014 01:00Z <japhb> smls: That mockup looks really good.  Aside from my general feeling that the green has been a bit too chartreuse for a while, I think it's excellent work.  +1 to make it so.
14:28 yoleaux 16 Feb 2014 18:40Z <raydiak> smls: I think you did an awesome job, though I also would consider a different book icon.  also, the colors are still up in the air, so feel free to play more with them. many people don't like the current combinations, and I'm not very sold on them myself. make sure you get moritz++'s input too. smls++!
14:28 yoleaux 16 Feb 2014 18:41Z <raydiak> smls: as for the structure, layout, etc, my first impression is that it is quite well-considered :)
14:28 yoleaux 17 Feb 2014 03:01Z <raydiak> smls: I changed the colors today, in case that affects anything you are/will be working on
14:29 smls oh, feedback
14:29 timotimo smls! :D
14:30 smls Has raydiak done any more doc.perl6.org style related work in the meantime?
14:33 smls I think I'll have a go at implementing those mockups in HTML/CSS this week, and while I'm at it try to make it look a bit more professional too (cf. the book icon and menu).
14:33 timotimo much appreciated!
14:33 timotimo i'm glad to see all areas of our documentation (at least those that exist already, excluding perl6-book) are getting some love right now
14:34 timotimo lue: moarvm is known to be horribly, horribly, horribly slow at concatenating strings
14:34 smls about the colors, I actually didn't mind the previous green/yellow scheme
14:34 smls but we can always have some more color bikeshedding later, once it's in HTML/CSS ;P
14:35 smls on a completely unrelated note:
14:35 smls perl6: my $a = 5; my $b = \$a; $$b = 10; say $a
14:35 camelia niecza v24-109-g48a8de3: OUTPUT«10␤»
14:35 camelia ..rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70: OUTPUT«5␤»
14:36 smls ^^ do Capture's no longer provide P5 reference-like syntrax for simple cases?
14:36 smls how would I write that instead?
14:36 FROGGS perl6: my $a = 5; my $b := $a; $b = 10; say $a
14:36 camelia rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70, niecza v24-109-g48a8de3: OUTPUT«10␤»
14:37 FROGGS even one char shorter :o)
14:37 timotimo https://www.youtube.com/watch?v=85LJh4sFi_M   -  oooh beautiful and stunning!
14:37 smls ok
14:38 smls what are Capture's used for now?
14:39 timotimo they represent the result of applying a list of things to a signature, iiuc
14:39 smls this doc suggest they're basically references on steroids: http://search.cpan.org/~lichtkind/Perl6-Doc-0.36/lib/Perl6/Doc/FAQ/Capture.pod
14:39 smls but it seems outdated
14:39 timotimo This ReleasePerl6-Doc-0.36 [Download] [Browse] 19 Mar 2008
14:39 timotimo Latest ReleasePerl6-Doc-0.47 [Download] [Browse] 05 Dec 2010
14:40 timotimo the current release doesn't contain the Capture FAQ document any more
14:40 salv0 joined #perl6
14:40 smls ok
14:40 timotimo but even that one is pretty outdated
14:41 timotimo perl6 has been moving a lot since 2010
14:42 timotimo i wonder at what point a LTS release of rakudo would be interesting
14:46 nwc10 how L is the Long in LTS?
14:46 nwc10 It's been talked about privately for Perl 5. I don't think that it works. All the distros use different dates for their own LTSes
14:47 nwc10 so either you'll never get them actually syncing to the release you'd like them to be LTS
14:47 virtualsue joined #perl6
14:47 nwc10 or you just add more lag on distribution update, because they aren't using this year's, but the upstream LTS from 18 months ago
14:48 [Coke] Parrot tried having LTS releases, and I think there was a disconnect between bundlers and developers. And rakudo is *still* changes often enough that I wouldn't recommend it until we hit 6.0.0"
14:49 [Coke] er, I was trying to edit that to "Perl 6 1.0"
14:50 shinobicl joined #perl6
14:50 timotimo mhm
14:51 nwc10 I think no sooner than 6.0.0
14:51 nwc10 and quite possibly give it a year or so to shake down after that
14:51 timotimo the idea would be to only merge performance improvements and obvious bugfixes, but no spec-change-related things or stuff that likely breaks compatibility with user's programs
14:51 masak what [Coke]++ said.
14:52 moritz atm it would also imply LTS versions of panda and essential modules
14:52 timotimo aye
14:52 moritz and you really don't want to do that right now :-)
14:52 nwc10 I thought that the *design* of Perl 6 was that the language was lexically modifiable
14:52 timotimo yes, i agree that "right now" is much too soon
14:52 timotimo in half a year, though ... ;)
14:53 nwc10 meaning that the parser can easily support more than one version of the language concurrently
14:53 timotimo oh. like ... "use backwardscompatible '2014.06'"?
14:53 nwc10 I thought it was more use v6.1.0
14:53 nwc10 or at least, v6.0.0 was meaning "keep me on the v6.0.0 spec"
14:54 timotimo i completely forgot about that feature for a moment
14:57 rurban joined #perl6
15:01 [Coke] I wonder if we need spec version/implementation version.
15:01 [Coke] spec version cannot be sufficient to say what code you want to run, I imagine.
15:02 vendethiel timotimo: are you boldly guessing a perl6.0.0 release date ? :d
15:02 timotimo :P
15:07 vendethiel joined #perl6
15:10 ajr joined #perl6
15:15 _thou joined #perl6
15:19 zakharyas joined #perl6
15:41 pmurias joined #perl6
15:46 rurban joined #perl6
15:50 xinming joined #perl6
15:51 ajr_ joined #perl6
16:05 yihshih joined #perl6
16:08 rurban joined #perl6
16:09 zakharyas joined #perl6
16:20 treehug88 joined #perl6
16:23 nebuchadnezzar joined #perl6
16:25 mtk joined #perl6
16:27 takadonet joined #perl6
16:27 takadonet morning
16:28 dmol joined #perl6
16:29 rurban joined #perl6
16:30 telex joined #perl6
16:38 clsn joined #perl6
16:42 [Coke] timotimo: (bug triage) and yet it never ends.
16:42 smls joined #perl6
16:43 * [Coke] ponders a mini doc on helping out with the bug queue.
16:43 [Coke] (in the mean time, feel free to ping me if you want to help)
16:43 zakharyas joined #perl6
16:43 [Coke] you==not timotimo specifically.
16:44 BenGoldberg joined #perl6
16:44 [Coke] the XSS error happens when you are NOT logged into RT, I think.
16:45 [Coke] nwc10: do you know what fivers do with their spam tickets? Is it something other than clicking the big S?
16:45 * [Coke] is happy to hit their recent spam tickets also when he's in there.
16:46 nwc10 pretty sure it's "click the big S"
16:46 nwc10 and yes, clicking S is appreciated
16:46 nwc10 kid51 is usually the most active on this
16:47 [Coke] nwc10: done.
16:58 yihshih__ joined #perl6
16:58 kshannon joined #perl6
16:58 segomos joined #perl6
17:03 xenoterracide joined #perl6
17:05 TimToady p6: say [1,2] X~ <a b>
17:05 camelia niecza v24-109-g48a8de3: OUTPUT«1 2a 1 2b␤»
17:05 camelia ..rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70: OUTPUT«1a 1b 2a 2b␤»
17:06 lizmat joined #perl6
17:06 breinbaas joined #perl6
17:06 mavcunha_ joined #perl6
17:06 FROGGS joined #perl6
17:06 revdiablo joined #perl6
17:06 ruoso_ joined #perl6
17:06 sftp joined #perl6
17:06 darutoko joined #perl6
17:06 klapperl joined #perl6
17:06 pnu joined #perl6
17:06 zamolxes_ joined #perl6
17:06 tokuhirom joined #perl6
17:06 aindilis joined #perl6
17:06 itz joined #perl6
17:06 ribasushi joined #perl6
17:06 nwc10 joined #perl6
17:06 sunnavy joined #perl6
17:06 pochi_ joined #perl6
17:06 erkan joined #perl6
17:06 dylanwh joined #perl6
17:06 Shozan joined #perl6
17:06 clsn joined #perl6
17:06 BenGoldberg joined #perl6
17:06 telex joined #perl6
17:06 _thou joined #perl6
17:06 salv0 joined #perl6
17:06 colomon joined #perl6
17:06 yoleaux joined #perl6
17:06 avuserow3 joined #perl6
17:06 eternaleye joined #perl6
17:06 exodist_ joined #perl6
17:06 BooK_ joined #perl6
17:06 JimmyZ joined #perl6
17:06 yath_ joined #perl6
17:06 bjz joined #perl6
17:06 go|dfish joined #perl6
17:06 rurban_ joined #perl6
17:06 lee_ joined #perl6
17:06 TimToady I assume that one is still considered a rakudobug
17:06 kshannon_ joined #perl6
17:06 xinming_ joined #perl6
17:07 [Coke] I just saw that bug yesterday, yes.
17:07 [Coke] https://rt.perl.org/rt3/Ticket/Display.html?id=74072
17:07 TimToady I assumed you were testing for still-bughood, but then you said something about test-neededs
17:07 TimToady so I wasn't sure
17:07 xfix joined #perl6
17:07 [Coke] ETOOMANYTESTS
17:08 [Coke] that one already has a test, even. t/spec/S03-metaops/cross.t
17:08 TimToady muchos gracias
17:08 [Coke] <nod>
17:09 TimToady hmm, was just looking at that code yesterday, maybe I can figure something out...
17:09 ggherdov joined #perl6
17:09 [Coke] m: say $]
17:09 camelia rakudo-moar 299d70: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/_cHikCl3Xjâ�¤Unsupported use of $] variable; in Perl 6 please use $*PERL_VERSIONâ�¤at /tmp/_cHikCl3Xj:1â�¤------> [32msay [33mâ��[31m$][0mâ�¤    expecting any of:â�¤        argument listâ�¤     …»
17:11 timotimo $PERL_VERSION?
17:11 timotimo not $*PERL<version>?
17:12 [Coke] https://rt.perl.org/Ticket/Display.html?id=120996 is a dupe, merged it.
17:16 pmurias joined #perl6
17:19 telex joined #perl6
17:29 japhb_ joined #perl6
17:37 bluescreen100 joined #perl6
17:37 vendethiel r: my $*a = 1; sub b { $*a = 2; c(); }; sub c { $*a = 3; say "c: {$*a.perl}"; }; say $*a.perl; b(); say $*a.perl;
17:37 camelia rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70: OUTPUT«1␤c: 3␤3␤»
17:37 vendethiel uw0t
17:38 Alina-malina joined #perl6
17:39 timotimo why does this confuse you?
17:40 takadonet left #perl6
17:40 TimToady if you want the subs to establish their own dynamic scope, you have to say 'my $*a = 3'
17:41 TimToady or 'temp $*a;'
17:41 vendethiel I'd expect the subs not to run before I call them
17:42 vendethiel as opposed to, say, a simple block
17:43 geekosaur ... what
17:44 hoverboard joined #perl6
17:44 nebuchadnezzar joined #perl6
17:44 17SAAJWDK joined #perl6
17:44 kurahaupo_mobile joined #perl6
17:44 dmol joined #perl6
17:44 mtk joined #perl6
17:44 dakkar joined #perl6
17:44 ggoebel1117 joined #perl6
17:44 raydiak joined #perl6
17:44 Yappo__________7 joined #perl6
17:44 ingy1 joined #perl6
17:45 segomos vendethiel: it ran in the order you called for
17:45 spider-mario joined #perl6
17:45 vendethiel geekosaur: is this a question of some sorts?
17:45 vendethiel I'm just blind, apparently
17:45 vendethiel definitely
17:46 geekosaur I cannot figure out why you think it ran before you called them
17:46 segomos it output 1, you called b which called c and output 'c: 3' and then output 3
17:46 vendethiel geekosaur, segomos : [18:45] <vendethiel> I'm just blind, apparently # my bad
17:46 segomos me too
17:47 vendethiel r: my $*a = 1; sub b { my $*a = 2; c(); }; sub c { $*a = 3; say "c: {$*a.perl}"; }; say $*a.perl; b(); say $*a.perl;
17:47 camelia rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70: OUTPUT«1␤c: 3␤1␤»
17:47 vendethiel i liek it
17:47 segomos interesting
17:48 vendethiel expected
17:48 colomon joined #perl6
17:49 segomos that is similar to the 'this' confusion people have in javascript
17:49 denis_boyun joined #perl6
17:51 masak not really, no.
17:52 masak well, depends on your threshold for similarity, I guess :P
17:53 jnthn Mars looks close tonight, so I'll svn ci
17:53 jnthn uh, wrong window
17:53 masak segomos: the main confusion people have about 'this' in JavaScript is that each individual routine invocation gets its own 'this', and so you cannot simply use 'this' from an inner function in a method, for example.
17:53 masak jnthn: *lol*
17:53 segomos it's scoping $*a to 'sub b', its caller.  in js people have confusion over 'this' being scoped to whatever called the function (which could be anything)
17:53 jnthn ...boy must that one look weird without context :)
17:53 masak jnthn: don't you dare try to explain!
17:53 masak jnthn: it's funnier this way :P
17:53 tadzik jnthn: ...what is a right window for that? :D
17:53 segomos lol
17:54 jnthn I think this is the first time lag has turned out to be a positive :P
17:54 masak segomos: no, 'this' is still scoped to the callee in JS.
17:54 masak segomos: are you sure you have the right definition of 'caller' in you mind?
17:55 masak segomos: or maybe you're completely misunderstanding how dynamic lookup works -- I cannot tell.
17:59 FROGGS joined #perl6
18:00 dalek specs: 5ef5f5c | larry++ | S32-setting-library/Containers.pod:
18:00 dalek specs: Add Range.sample for TheDamian++
18:00 dalek specs: review: https://github.com/perl6/specs/commit/5ef5f5ca42
18:01 [Coke] docs.perl.org - it would be nice if you got an indication that a search returned no results.
18:01 moritz raydiak: ^^
18:01 [Coke] er, http://doc.perl6.org/
18:02 [Coke] also, I'd like to be able to hit "ord<enter>" in the search box and have it go somewhere.
18:03 moritz TimToady: it seems your last spec commit added much more than Range.sample
18:03 moritz TimToady: like, duplicated the Buf section
18:03 segomos masak: no i get what you're saying, but 'this' in js isn't always equal to .caller
18:03 TimToady oops
18:04 masak segomos: there is some weapons-grade confusion going on here. why would 'this' be equal to .caller?
18:04 segomos i'm generalizing the way the scope of a variable works
18:04 SamuraiJack_ joined #perl6
18:05 masak yes, that is probably where we need to begin.
18:05 segomos masak: i think because i'm half assedly paying attention -
18:05 masak segomos: 'lexical lookup' vs 'dynamic lookup' -- do these terms sound familiar?
18:05 segomos yea
18:05 masak segomos: well, the thing with 'this' is that it does neither. it's simply invocation-specific. that is, the *invocation* sets 'this'.
18:06 masak sometimes to a sensible object (like with methods and constructors), sometimes to the global object (like with functions), and sometimes to something you specify (like with call/apply).
18:06 berekuk joined #perl6
18:06 segomos yea i'm familiar with how it works
18:06 masak ok.
18:07 masak then I was hoping you would also say that it's nothing like dynamic lookup :)
18:07 masak food &
18:07 dalek specs: 73c60fc | larry++ | S32-setting-library/Containers.pod:
18:07 dalek specs: remove accidental duplication
18:07 dalek specs: review: https://github.com/perl6/specs/commit/73c60fc5fe
18:07 pmurias joined #perl6
18:08 molaf joined #perl6
18:09 telex joined #perl6
18:11 segomos masak: ahh :) sorry, I'm getting frustrated with this SAP soap thing and only half reading (apparently)
18:14 synopsebot joined #perl6
18:14 FROGGS segomos: I'll fell the joy (pain, really) of SAP soap in near future too :o)
18:19 masak segomos: s'ok. I wasn't sure which one of us was confused, to be honest.
18:19 * moritz was confused
18:19 moritz :-)
18:22 ajr joined #perl6
18:33 xragnar joined #perl6
18:39 tgt joined #perl6
18:42 khisanth_ joined #perl6
18:44 ajr__ joined #perl6
18:46 pmurias joined #perl6
18:53 segomos FROGGS: ahh, i think i just got BOBJ QAAWS working which means i can do web type development instead of using these hokey tools
18:57 FROGGS segomos: my project will probably start in april or may, before that I'wont know what I am supposed to do
18:59 SamuraiJack_ joined #perl6
19:02 nwc10 https://www.google-melange.com/gsoc/org2/google/gsoc2014/tpf
19:02 treehug8_ joined #perl6
19:03 timotimo we wanted to get the peephole optimizer task removed from the ideas page
19:03 timotimo can someone do that?
19:04 nwc10 surely it's a wiki, so anyone could?
19:04 timotimo ... i suppose at least registration is required?
19:04 timotimo i thought there'd probably be a requirement to hve some privileges
19:06 kurahaupo_mobile joined #perl6
19:19 Rotwang joined #perl6
19:25 lue timotimo: interesting. I was looking at join last night thinking to myself "kcachegrind says it's only called 4 times, it can't be what's *really* slowing everything down".
19:26 timotimo http://t.h8.lv/p6bench/2014-02-20-rakudo.html ← have a closer look
19:27 timotimo at some point we ought to have concurrency benchmarks, too :)
19:31 denis_boyun__ joined #perl6
19:31 lue You should probably leave parrot out of those, unless you want a depressing flatline on all the charts :)
19:32 * masak .oO( this... is a flatlined Parrot! )
19:32 timotimo parrot still beats out moarvm when it comes to string operations
19:32 [Coke] On the contrary, don't we want stats on parrot vs. moar?
19:32 grondilu moritz: in https://github.com/moritz/perl6-wtop/blob/master/practices.pod, you could possibly mention infix:<**> as an other exception to the rule about putting spaces around an infix operator.
19:32 timotimo and otherwise it's usually only like 2x faster
19:32 nwc10 I thought that too
19:32 lue [Coke]: not for concurrency I don't think :)
19:32 nwc10 and what's curious is why parrot wins on while_concat
19:32 timotimo it's not curious
19:33 nwc10 OK. Why isn't it curious?
19:33 timotimo moar's way of concatenating strings is absolutely abysmally pessimized
19:33 timotimo our rope implementation is thoroughly troubled
19:33 masak grondilu: you mean infix:<**> in the Regex slang?
19:34 grondilu no, I mean the exponentiation operator.
19:34 lue jnthn mentioned just recently that the ropes are forced into being simple strings (because the ropey parts of the impl. are broken), which is apparently absolutely terrible for performance.
19:34 grondilu it's an infix, isn't it?
19:34 timotimo it's doing the pessimized "copy all the things" thing
19:35 timotimo and also, we're forcing all strings to use 32bit codepoint storage even if it would fit into 8bit
19:35 timotimo i think parrot is better at that very point than moar
19:36 jnthn I don't think the 32-bit/8-bit is the big deal.
19:36 jnthn It's the building up an intricate data structure, only to throw it away again.
19:36 timotimo how about the way we're accessing the codepoints for copying?
19:37 jnthn timotimo: Yes, that is more the problem.
19:37 timotimo so, a fast path for two 32bit non-rope strings ought to be a bit of a help?
19:37 timotimo one that uses memcpy?
19:38 jnthn timotimo: Yes.
19:38 timotimo seems like an easy thing to whip up in a few minutes right now :)
19:38 kaare_ joined #perl6
19:38 [Coke] jnthn: do you have any need of curated tickets from RT to help you find things to work on, or are you (as I suspect) all set in the find stuff to do area?
19:40 jnthn [Coke]: I've got a fairly sizable bunch of things to do in the near future.
19:40 kurahaupo_mobile joined #perl6
19:40 jnthn [Coke]: At least one RT ticket being one of the most urgent.
19:41 [Coke] hokay.
19:43 [Coke] Where did the node.js backend stuff end up?
19:43 panchiniak joined #perl6
19:45 FROGGS jnthn: what is the sanest way to access a BOOThash on jvm?
19:45 FROGGS nqp::backendconfig()<foo> does not work
19:46 FROGGS "No such method 'at_key' for invocant of type 'BOOTHash'"
19:46 jnthn FROGGS: From perl 6 or nqp?
19:46 jnthn Oh, then Perl6
19:46 jnthn Well, nqp::atkey(.., 'foo')
19:46 FROGGS in rakudo's core_epilog.pm to be exact
19:46 FROGGS hmmm, I think I tried that....
19:46 jnthn Try it righter
19:46 jnthn :P
19:46 FROGGS yeah :P
19:48 dalek rakudo-star-daily: 5029d43 | coke++ | log/ (5 files):
19:48 dalek rakudo-star-daily: today (automated commit)
19:48 dalek rakudo-star-daily: review: https://github.com/coke/rakudo-star-daily/commit/5029d43dd7
19:48 FROGGS [Coke]: btw, I tried to fix that EVAL for pugs... but I gave up after a while
19:51 FROGGS jnthn: seems to work(-ish), there are more accesses that need to be nqp::atkey, likely that I've just not seen that the first change succeeded
19:54 [Coke] FROGGS: me too, which is why I opened a ticket. :)
19:54 FROGGS yeah
19:54 [Coke] eval looks special. (other stuff, it was a simple mod to Prim.hs)
19:54 FROGGS :/
19:54 * [Coke] casts "Summon au"
19:54 FROGGS expecto au?
20:00 tgt joined #perl6
20:01 rindolf joined #perl6
20:05 masak you can only cast au if the thought you're holding is happy enough.
20:06 timotimo %)
20:08 pdcawley joined #perl6
20:09 kivutar joined #perl6
20:09 FROGGS *g*
20:09 FROGGS so, flying on a broomstick doesn't ount, ehh?
20:09 FROGGS count*
20:09 FROGGS I need to reparo my keyboard -.-
20:09 * masak .oO( we all knew that au was brilliant, but we hadn't realized until now she's a patronus )
20:17 masak pmurias: another thing that's usually written in all-caps: heredoc terminating strings.
20:18 masak by the way, it's an interesting language evolution that heredoc is spelled "here-doc" in perldoc.
20:18 tgt joined #perl6
20:26 Sqirrel joined #perl6
20:29 timotimo lue: i have a moarvm patch that you can try with your pod6 htmlify stuff - shouldn't need to re-build rakudo or nqp or the module.
20:29 lue I'd be willing to try it. Do I need latest MoarVM?
20:30 timotimo it's based on MoarVM master, but it should apply cleanly to older MoarVM as well
20:30 timotimo the branch you want to use is flatten_fastpath in moarvm
20:31 lue oh, so I can just switch to it then, no manual patch application like I was thinking :) .
20:31 timotimo that's right
20:31 timotimo i figured that'd be much nicer for everyone
20:35 timotimo can't tell if the spectests are clean yet.
20:40 lue O.O before: about 1246.0347964 seconds  after: 244.9768101 seconds. (Most of that is hung up in File::Find::find; all steps after the last use of find() blinked past before I knew what was happening :D)
20:40 lue timotimo++ ! \o/
20:40 lue Huh, all of a sudden I get links pointing to https://github.com/perl6/roast/blob/master//S17-async/syntax.t#L33\  (notice the backslash) can't tell if that's me or the MVM patch.
20:40 timotimo sounds good
20:41 timotimo hm. interesting.,
20:41 FROGGS interesting\
20:41 FROGGS :P
20:41 timotimo could you be so kind as to golf it?
20:42 lue I'll revert to pre-patch moar, and see if it comes up again.
20:43 lue (that'll take about 15 minutes to run through though)
20:43 timotimo hah
20:43 timotimo can you reduce the workload?
20:44 lue Yes, actually. I'll just have to copy over a single synopsis to use.
20:44 timotimo spectests are clean, annoyingly
20:45 timotimo so if it's my patch that's to blame, we could really use a spectest
20:45 timotimo well, a regression test really
20:47 lue *headdesk* "Obtained 0 synopses." well, no wonder it went by so fast :P  This also means the backslashes are my fault, almost surely.
20:47 timotimo huh?
20:47 timotimo so the performance measurement was improper, too?
20:48 lue I forgot to switch back to the pod6ify branch of specs before testing you patch. Sorry :(
20:48 timotimo hehehe
20:48 timotimo that's fine. it should still give improved performance
20:48 timotimo i'll run the proper benchmarks soon
20:49 lue The backslash is because, as I thought, it's a leftover from moving from double quotes to a heredoc :)
20:50 lue (link still points to the correct line and file, interestingly)
20:51 masak lue: did you know heredocs can be double-quoted, too? :)
20:52 lue what do you mean?
20:52 masak qq:to/FOO/
20:52 lue ah, I just meant a literal "" :)
20:52 masak yeah, I figured.
20:53 zaydon joined #perl6
20:54 mcglk joined #perl6
21:04 araujo joined #perl6
21:13 lue timotimo: 1246.0347964 seconds  vs.  1222.1842851 seconds (can't tell if that's significant; a previous pre-patch run was at 1237.8338873)
21:14 timotimo ho-hum.
21:16 timotimo in that case, something else should be a bit slower
21:16 timotimo could be that you use join and that doesn't hit the fast path
21:16 araujo joined #perl6
21:17 timotimo and string_repeat doesn't usually hit the fast path either
21:17 lue I think I tossed out my kcachegrind stuff at some point, couldn't make much of it anyway. [at least, *I* couldn't :)]
21:19 timotimo i have the same problem
21:19 d4l3k_ joined #perl6
21:20 Pleiades` joined #perl6
21:20 lue I think at some point it's just going to take time processing the synopses, though hopefully less than the 10 or so minutes it does currently take.
21:22 lue Oh, dur. I've been using perl6-p for the profiling stuff. If perl6-m actually had profiling support, I'd get actually useful data from there :) .
21:22 timotimo you can still cachegrind that :P
21:23 lue I tried using callgrind once, IIRC it was useless (could try again though)
21:23 timotimo yeah :(
21:24 timotimo diakopter has been getting opinions/input on what people need for a profiler for moarvm, so at some point he'll have something for us to work with
21:29 lue I just want something that'll show off bottlenecks in code. Not sure how useful that suggestion is though :)
21:29 timotimo lue: can you tell if you're using join more often than concatenate?
21:29 lue If my attempt to run callgrind on perl6-m --doc=HTML fails, I'll profile perl6-p and tell you.
21:31 timotimo thank you
21:32 timotimo how often MVM_string_repeat happens would also be interesting
21:32 timotimo those would all require individual fastpaths unfortunately
21:32 lue maybe to the point where a better string impl is needed, perhaps? :)
21:33 timotimo yes.
21:33 timotimo please make a better string impl if you can :P
21:33 lue I just read about ropes not too long ago (to better know the more technical bits), so I don't know how good I'd be :P
21:34 virtualsue joined #perl6
21:35 lue This perl6-m callgrind might take a while :)
21:35 timotimo yeah, probably :(
21:36 lue Perhaps I shouldn't've picked S02, the largest of the synopses...
21:36 timotimo heh :)
21:37 lue It takes ≈1m30s on perl6-m, so add into that the 20..50x slowdown I read valgrind taking, and I expect at least 10 more minutes to go into this.
21:39 masak r: "foo" ~~ /(bar)? foo/; say ~$0
21:39 camelia rakudo-jvm 299d70, rakudo-moar 299d70: OUTPUT«use of uninitialized value of type Any in string context␤␤»
21:39 camelia ..rakudo-parrot 299d70: OUTPUT«use of uninitialized value of type Any in string context  in block  at /tmp/tmpfile:1␤␤␤»
21:39 masak r: "foo" ~~ /(bar)? foo/; say $0.^name
21:39 camelia rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70: OUTPUT«Any␤»
21:39 masak r: "foo" ~~ /(bar)? foo/; say ~($0 // "")
21:39 camelia rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70: OUTPUT«␤»
21:39 timotimo hm, should that be Nil?
21:42 moritz r: say so "foo" ~~ /(bar)? foo/;
21:42 camelia rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70: OUTPUT«True␤»
21:44 d4l3k_ joined #perl6
21:45 lue timotimo: trying my hand at ropes could possibly be interesting, but not only do I need to be familiar with the MVM codebase, I also would much prefer experimenting in a separate, useless project first :) [Also, if I must use C, that's another thing...]
21:45 timotimo you'd have to be using C, aye
21:45 timotimo and the code for ropes is kinda complicated
21:46 timotimo it uses the visitor pattern to work on ropes, which is pretty nice
21:46 timotimo but it's not obvious how exactly things work
21:46 timotimo and especially where the gremlins are
21:47 lue Not saying I couldn't use C, just that it'd take some time to beat the C++11 outta me :P
21:48 timotimo hah
21:48 timotimo also you can't use C99
21:48 awwaiid joined #perl6
21:49 lue Yeah, I caught that in #moarvm. That means none of that mixing declarations and code I got bitten by in some project I had to repair a while ago, right?
21:50 timotimo yeah, declaration has to be at the beginning of blocks :(
21:50 timotimo i've made that mistake quite often
21:51 lue Good thing I'm not in charge of moar, else I'd say "use a better compiler" :P
21:51 timotimo hah
21:51 timotimo msvc++ could in theory be "better"
21:52 timotimo allegedly, MS's response to "MSVC is lagging far behind the standards!" is "use msvc++ instead, dummies"
21:54 moritz can anybody tell me what's wrong with this table?
21:54 moritz https://github.com/perl6/doc/blob/master/lib/regexes.pod#L253
21:54 moritz it's not rendered as a table at all, neither by Pod::To::Text nor by Pod::To::HTML
21:55 timotimo does it get confused by initial empty lines?
21:56 moritz removing it doesn't unconfuse it
21:56 timotimo also, i think the table above that for the short and long unicode thingies ought to have many more empty lines
21:56 moritz let's start with the simpler one :-)
21:56 timotimo since i believe seeing at least one empty line in between puts it into "one row per paragraph" mode rather than "one row per line" mode
21:57 masak 'night, #perl6
21:57 timotimo 'night masak
21:57 lue ♞ masak o/
21:57 lue moritz: did you remove the final empty line too?
21:57 moritz lue: yes
21:57 moritz fwiw the table in lib/Whatever.pod works fine
21:58 moritz and it also has newlines before and after the table body
21:59 timotimo hmm
21:59 moritz aaaargh
21:59 lue Oh! Try wrapping | and + in V<>s ! (may need :allow<V> on the table)
22:00 moritz if I add an 'a' before each row "fixes" it
22:00 timotimo does :allow actually work in tables now?!
22:00 lue maybe -
22:00 lue timotimo: it does in code blocks at least, not sure if it works everywhere though
22:00 zakharyas joined #perl6
22:00 lue (theoretically every cell is a nested =para, so you shouldn't need :allow)
22:00 lue *maybe - too
22:00 moritz what is V<>?
22:01 timotimo "verbatim"
22:01 timotimo lue: not in the implementation i've seen :)
22:01 moritz how does it differ from C<>?
22:01 colomon joined #perl6
22:01 timotimo doesn't change formatting
22:01 moritz when I add V<...> around the first column, it's rendered as a table, including the V<>
22:02 lue the point is to not have whitespace on at least one side of the symbol, because they're looked at as cell dividers here.
22:02 timotimo yes, as i said, formatting codes inside tables are sadly NYI
22:02 timotimo really sad about that :(
22:02 timotimo so maybe use "+" instead?
22:02 timotimo that'll be rendered for "humans", but it at least won't look weird
22:02 moritz so, currently there's no way to render a table that has | as a column?
22:03 moritz timotimo: use + where?
22:03 lue not one surrounded by whitespace at least
22:03 timotimo V<...> would counter that problem, but is NYI
22:03 timotimo moritz: use \" around the operators
22:04 * lue thinks Pod6 tables need some refactoring in general; there're too many styles to say "Pod6 will figure it out, don't worry"
22:05 tadzik that's how I wrote it, heh
22:05 tadzik timotimo has a much saner implementation
22:05 timotimo somewhat
22:05 tadzik it's interpreting S26 differently, but correcter I think
22:05 timotimo tadzik: did you see i showed it to Mouq++ and he liked it?
22:05 moritz timotimo: then I have \" around the operators in the output :(
22:05 tadzik timotimo: didn't notice, no
22:06 timotimo moritz: yes, you will. but it won't be as confusing to the user as having a visible V<..> around it
22:06 timotimo hm. how do &foo; get treated there?
22:06 timotimo would that perhaps work as a blatant hack?
22:06 moritz it's not *that* kind of an operator
22:07 moritz but rather regex unicode class combination operators
22:07 moritz as in <:alpha+:digit>
22:07 timotimo i know.
22:07 moritz found a hack!
22:07 moritz using non-breaking spaces around the operator :-)
22:07 timotimo hah, clever
22:07 moritz a sad kind of clever
22:07 timotimo agreed
22:08 dalek doc: fbc8306 | moritz++ | lib/regexes.pod:
22:08 dalek doc: [regexes] fix tables / work around rakudo limitations
22:08 dalek doc: review: https://github.com/perl6/doc/commit/fbc83068ab
22:09 lue .oO( =begin table :tablespec<+-+|+-+|+++++>  / :tablespec<┏━┓┃┛━┗┃┳┫┻┣╋> ...)
22:09 timotimo %)
22:32 Mouq joined #perl6
22:34 Mouq moritz: "so, currently there's no way to render a table that has | as a column?" :/ sorry
22:35 Mouq moritz: It's also a problem I've noticed e.g. there's no way to do L<infix:<(\|)> | infix ∪ > or whatever
22:35 Mouq The parsing just needs to be better
22:36 lue timotimo: I just got callgrind output for perl6-m --doc=HTML S02-bits.pod6 (took just over 1h), and I got a lot of stuff on MVM, but it looks like no perl6-specific stuff
22:37 Mouq I won't have time tonight, but I'm working on getting timotimo's table implementation into Rakudo
22:37 Mouq Actually, I may as well commit and push to a branch
22:37 Mouq It at least builds
22:37 lue Mouq: in that case, I don't think you space around the L<> bar
22:38 lue *you need space
22:39 Mouq lue: The problem is that the Pod::To::* modules have to parse the inside of L<>, and that's stupid
22:39 lue oh, is that what the \| is for?
22:40 Mouq L<> should produce a Pod::FormattingCode(type<L>, link<whatev>, content([Pod::*]))
22:40 dmol joined #perl6
22:41 rurban joined #perl6
22:41 Mouq This isn't something that S26 says, but I can't imagine an argument to the contrary
22:42 lue Mouq: would L<infix:<(\|)>|infix ∪> break things?
22:43 Mouq lue: With Pod::To::HTML it produces content => 'infix:<(\', link => ')>|infix ∪'
22:44 lue couldn't you use V<|> or E<0x7C> where needed?
22:45 Mouq Unfortunately no.
22:45 Mouq Not with the way things are now, anyway.
22:46 Mouq I mean
22:46 pippo joined #perl6
22:46 Mouq yes, that would be the way to do it, rather than \|
22:46 Mouq but it still wouldn't work
22:50 dalek rakudo/pod-table: 25761c4 | Mouq++ | src/Perl6/ (2 files):
22:50 dalek rakudo/pod-table: WIP, converting timotimo++'s pod-table parser to the necessary NQP
22:50 dalek rakudo/pod-table: review: https://github.com/rakudo/rakudo/commit/25761c4814
22:50 lue not even E<> ? That's unfortunate.
22:57 Mouq r: say "\c[NL]"
22:57 camelia rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/tmpfile�Unrecognized character name NL�at /tmp/tmpfile:1�------> [32msay "\c[NL[33m�[31m]"[0m�»
22:57 Mouq r: say "\c[NEL]"
22:57 camelia rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/tmpfile�Unrecognized character name NEL�at /tmp/tmpfile:1�------> [32msay "\c[NEL[33m�[31m]"[0m�»
22:58 Mouq r: say "\c[INTERROBANG]"
22:58 camelia rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70: OUTPUT«‽␤»
23:03 Mouq r: my $code = 'INTERROBANG'; say nqp::codepointfromname($code)
23:03 camelia rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70: OUTPUT«8253␤»
23:04 Mouq r: my $code = 'INTERROBANG'; say chr(nqp::codepointfromname($code))
23:04 camelia rakudo-parrot 299d70, rakudo-jvm 299d70, rakudo-moar 299d70: OUTPUT«‽␤»
23:04 Mouq Any other way to do this?
23:06 pdcawley joined #perl6
23:06 lue r: say nqp::namefromcodepoint(8253);
23:06 camelia rakudo-jvm 299d70: OUTPUT«===SORRY!===␤No such method 'jast' for invocant of type 'NQPMu'␤»
23:06 camelia ..rakudo-parrot 299d70, rakudo-moar 299d70: OUTPUT«===SORRY!===␤Error while compiling op namefromcodepoint: No registered operation handler for 'namefromcodepoint'␤»
23:06 lue Aw :)
23:06 treehug88 joined #perl6
23:15 colomon joined #perl6
23:20 vendethiel joined #perl6
23:45 skids joined #perl6
23:47 hoverboard joined #perl6

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo